Do I need to write code to create a diagram?

No, you can describe your diagram in natural language, such as 'Transformer encoder-decoder architecture with multi-head attention,' and the AI handles the layout, labeling, and rendering automatically.

What document types are supported for quality thresholds?

It includes pre-set thresholds for journals (8.5/10), conference papers (8.0), theses, grant proposals, preprints, reports, posters (7.0), and presentations (6.5).

Where are the generated diagrams stored?

All diagrams are automatically saved in the figures/ subfolder of your local repository, making them easy to reference in LaTeX files, Markdown papers, or slide decks.

How does the quality review system work?

The skill uses Gemini 3 Pro to evaluate generated diagrams against 10-point criteria including scientific accuracy, clarity, and label quality. If a score falls below the set threshold for your document type, it automatically triggers a refined regeneration.

Scientific Schematics

Name: Scientific Schematics
Author: pur3v4d3r

bypur3v4d3r

•

Data Science & ML

Generates publication-quality scientific diagrams and neural network architectures with AI-powered quality review.

Scientific Schematics automates the creation of complex visual representations for academic and technical publications. By leveraging Nano Banana Pro AI for image generation and Gemini 3 Pro for rigorous quality assessment, the skill produces professional-grade flowcharts, biological pathways, and system architectures. It features a unique smart iteration system that evaluates diagrams against specific document thresholds—such as Nature-level journals or conference posters—ensuring the output meets high scientific standards without manual refinement. This tool is ideal for researchers, data scientists, and engineers who need to transform technical concepts into visual assets ready for peer review or presentations.

Key Features

01Automated quality review using Gemini 3 Pro against document-type thresholds

02Smart iterative refinement to improve diagrams until they meet publication standards

03AI-powered generation of complex neural network architectures and system diagrams

041 GitHub stars

05Support for diverse scientific formats including CONSORT flowcharts and biological pathways

06Publication-ready styling including colorblind-friendly palettes and professional typography

Use Cases

01Creating standardized participant flow diagrams for clinical studies and methodology sections

02Visualizing complex biological signaling pathways or chemical interactions for academic posters

03Generating high-resolution architecture diagrams for AI/ML research papers

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add pur3v4d3r/pur3-pkb-codebase scientific-schematics

For use in Claude.ai and ChatGPT

Download Skill